Abstract

The systematic reuse of components for software development is an increasingly popular way to improve productivity and quality in the creation of software. Companies moving towards a reuse-driven development process, are facing the trade-off between the increased cost of developing reusable code and the cost savings they can achieve with it. In order to turn the cost-benefit equation to their advantage, they have to minimize cost in two areas: the development cost of reusable components, and the retrieval cost of reusable components from the software repository. The latter can be pursued in several ways. The traditional approach to component retrieval is a detailed classification of all components with key terms and the use of search programs to find the appropriate component in the repository. The complex nature of components makes the identification of suitable key terms difficult. In this paper we are presenting an approach to overcome those difficulties by using an enterprise-level model as an alternative to component retrieval. To illustrate the practical feasibility of the alternative model, a specific implementation of this approach by a business process and accounting system developer is described.

Share

COinS